Ovarian cancer often goes undetected and untreated until advanced stages. Early detection of ovarian cancer could substantially improve the result of therapy and reduce mortality. Therefore, it is of prime importance to develop early diagnostic biomarkers and techniques. Tumor growth is largely dependent on sufficient vascularization. Ovarian tumors produce various angiogenic factors, such as VEGF, FGF, PDGF, IL-8, that stimulate new blood vessel formation providing oxygen and nutrients for tumor growth. We hypothesize that a broad panel of various angiogenic factors produced or induced by ovarian tumor could be predictive of early stage ovarian cancer. Our long-range goal is to establish a reliable blood-based assay for early detection of ovarian cancer. The main objective of this application is to identify a panel of serum angiogenic factors that are most prominent in ovarian cancer patients. To accomplish this objective, we will pursue two specific aims: 1. To screen serum of ovarian cancer patients using a multiplex array of ovarian cancer-associated angiogenic factors. In this specific aim we propose to analyze serum titers of VEGF, FGF, IL-8, IL-6, MCP-1, M-CSF, G-CSF, PDGF D, EGF and TGF-beta in ovarian cancer patients, patients with benign tumors, and healthy women, using LUMINEX technology. 2. To apply statistical modeling techniques to identify an integrated panel of angiogenic factors with superior diagnostic power. We will analyze the resultant data using newly developed multi-parameter analysis software to select panel of cytokines with high disease association. Using current methods in the area of classification, the clinical and cytokine data will be used to develop a classification rule that can be used to identify ovarian cancer patients. Our preliminary testing of five angiogenic factors using LUMINEX technology revealed significant differences in the levels of EGF, G-CSF, IL-8 and IL-8 between healthy women and patients with early stage ovarian cancer. The validity and fruitfulness of this approach for the development of early detection markers for ovarian cancer will be determined.